US12038608B2 - Self-aligning photonic interconnections for photonic integrated circuits - Google Patents
Self-aligning photonic interconnections for photonic integrated circuits Download PDFInfo
- Publication number
- US12038608B2 US12038608B2 US17/470,803 US202117470803A US12038608B2 US 12038608 B2 US12038608 B2 US 12038608B2 US 202117470803 A US202117470803 A US 202117470803A US 12038608 B2 US12038608 B2 US 12038608B2
- Authority
- US
- United States
- Prior art keywords
- waveguide
- plane
- waveguides
- pic
- photonic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active, expires
Links
- 238000010168 coupling process Methods 0.000 claims abstract description 97
- 238000005859 coupling reaction Methods 0.000 claims abstract description 96
- 230000008878 coupling Effects 0.000 claims abstract description 95
- 238000006073 displacement reaction Methods 0.000 claims abstract description 21
- 239000000758 substrate Substances 0.000 claims abstract description 16
- 229920000642 polymer Polymers 0.000 claims description 61
- 230000003287 optical effect Effects 0.000 abstract description 29
- 238000012546 transfer Methods 0.000 abstract description 26
- 238000000034 method Methods 0.000 abstract description 19
- 238000013459 approach Methods 0.000 abstract description 11
- 230000008569 process Effects 0.000 abstract description 8
- 230000003993 interaction Effects 0.000 abstract description 6
- 238000005516 engineering process Methods 0.000 abstract description 5
- 230000005693 optoelectronics Effects 0.000 abstract description 4
- 208000036758 Postinfectious cerebellitis Diseases 0.000 description 49
- 239000010410 layer Substances 0.000 description 40
- 239000011295 pitch Substances 0.000 description 26
- 239000000463 material Substances 0.000 description 22
- 229910052581 Si3N4 Inorganic materials 0.000 description 20
- HQVNEWCFYHHQES-UHFFFAOYSA-N silicon nitride Chemical compound N12[Si]34N5[Si]62N3[Si]51N64 HQVNEWCFYHHQES-UHFFFAOYSA-N 0.000 description 20
- 238000005253 cladding Methods 0.000 description 18
- 238000004519 manufacturing process Methods 0.000 description 15
- 230000006870 function Effects 0.000 description 14
- 230000005540 biological transmission Effects 0.000 description 12
- 238000003491 array Methods 0.000 description 11
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 10
- 230000008901 benefit Effects 0.000 description 10
- 238000003780 insertion Methods 0.000 description 10
- 230000037431 insertion Effects 0.000 description 10
- 230000010354 integration Effects 0.000 description 10
- 238000013461 design Methods 0.000 description 9
- 239000000835 fiber Substances 0.000 description 8
- 238000004088 simulation Methods 0.000 description 8
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 6
- 229920006254 polymer film Polymers 0.000 description 6
- 229910052710 silicon Inorganic materials 0.000 description 6
- 239000010703 silicon Substances 0.000 description 6
- 238000003384 imaging method Methods 0.000 description 5
- 239000000377 silicon dioxide Substances 0.000 description 5
- GPXJNWSHGFTCBW-UHFFFAOYSA-N Indium phosphide Chemical compound [In]#P GPXJNWSHGFTCBW-UHFFFAOYSA-N 0.000 description 4
- 230000007704 transition Effects 0.000 description 4
- 229910052681 coesite Inorganic materials 0.000 description 3
- 229910052906 cristobalite Inorganic materials 0.000 description 3
- 238000000206 photolithography Methods 0.000 description 3
- 229920003229 poly(methyl methacrylate) Polymers 0.000 description 3
- 239000004926 polymethyl methacrylate Substances 0.000 description 3
- 229910052682 stishovite Inorganic materials 0.000 description 3
- 229910052905 tridymite Inorganic materials 0.000 description 3
- ARXHIJMGSIYYRZ-UHFFFAOYSA-N 1,2,4-trichloro-3-(3,4-dichlorophenyl)benzene Chemical compound C1=C(Cl)C(Cl)=CC=C1C1=C(Cl)C=CC(Cl)=C1Cl ARXHIJMGSIYYRZ-UHFFFAOYSA-N 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- MPTQRFCYZCXJFQ-UHFFFAOYSA-L copper(II) chloride dihydrate Chemical compound O.O.[Cl-].[Cl-].[Cu+2] MPTQRFCYZCXJFQ-UHFFFAOYSA-L 0.000 description 2
- 238000001459 lithography Methods 0.000 description 2
- 230000001902 propagating effect Effects 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 239000000523 sample Substances 0.000 description 2
- 230000003746 surface roughness Effects 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 230000014616 translation Effects 0.000 description 2
- 239000004593 Epoxy Substances 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000004026 adhesive bonding Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000011109 contamination Methods 0.000 description 1
- 239000012792 core layer Substances 0.000 description 1
- 238000006880 cross-coupling reaction Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001839 endoscopy Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 125000003700 epoxy group Chemical group 0.000 description 1
- 230000001747 exhibiting effect Effects 0.000 description 1
- 229920005570 flexible polymer Polymers 0.000 description 1
- 239000004811 fluoropolymer Substances 0.000 description 1
- 229920002313 fluoropolymer Polymers 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000001093 holography Methods 0.000 description 1
- 239000005001 laminate film Substances 0.000 description 1
- 238000002844 melting Methods 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- 239000012528 membrane Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000465 moulding Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000003534 oscillatory effect Effects 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 238000002161 passivation Methods 0.000 description 1
- 238000000059 patterning Methods 0.000 description 1
- 238000005498 polishing Methods 0.000 description 1
- 150000003071 polychlorinated biphenyls Chemical class 0.000 description 1
- 229920000647 polyepoxide Polymers 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 235000012239 silicon dioxide Nutrition 0.000 description 1
- 229910000679 solder Inorganic materials 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
- 238000003892 spreading Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B6/00—Light guides; Structural details of arrangements comprising light guides and other optical elements, e.g. couplings
- G02B6/10—Light guides; Structural details of arrangements comprising light guides and other optical elements, e.g. couplings of the optical waveguide type
- G02B6/12—Light guides; Structural details of arrangements comprising light guides and other optical elements, e.g. couplings of the optical waveguide type of the integrated circuit kind
- G02B6/122—Basic optical elements, e.g. light-guiding paths
- G02B6/125—Bends, branchings or intersections
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B6/00—Light guides; Structural details of arrangements comprising light guides and other optical elements, e.g. couplings
- G02B6/10—Light guides; Structural details of arrangements comprising light guides and other optical elements, e.g. couplings of the optical waveguide type
- G02B6/12—Light guides; Structural details of arrangements comprising light guides and other optical elements, e.g. couplings of the optical waveguide type of the integrated circuit kind
- G02B2006/12133—Functions
- G02B2006/12147—Coupler
Definitions
- Interconnects are indispensable in electronics, enabling circuit components with diverse functionalities to be assembled together into complex systems.
- An optical interconnect should be single-mode, low loss, easy to manufacture, and compatible with high-density electrical interconnect technologies such as flip-chip bonding. Typically, this is achieved using tapered adiabatic couplers that interface waveguides on a photonic integrated circuit (PIC) to a common optical substrate.
- PIC photonic integrated circuit
- alternative approaches have been proposed in the past, including photonic “wirebonds” that connect PICs through flexible polymer waveguides, integrated optical microlenses with through-substrate vertical grating couplers, and bulk optical components such as parabolic reflectors microfabricated into polymer films. Small numbers of PICs can also be connected by conventional fiber with edge coupling or grating coupling, but this is challenging to scale to large numbers of channels.
- a self-aligning photonic interconnect technology that is insensitive to misalignment can be used to connect PICs and other devices with photonic components.
- This technology uses the interaction created by two waveguides crossing at an angle, giving rise to efficient evanescent coupling at their intersection.
- this coupler is invariant to translational misalignment, as two waveguides at an angle will still intersect after in-plane translation.
- the coupling efficiency of this structure is far more insensitive to angular misalignment than more conventional approaches such as edge coupling.
- a cantilevered self-aligning coupler also relaxes tolerances for out-of-plane misalignment.
- an inventive self-aligning photonic interconnect has several advantages over other PIC connectors. It can connect a single-mode waveguide on one PIC to a single-mode waveguide in the same PIC or another PIC. Its propagation loss can be extremely low, e.g., below 0.1 dB/cm. Its insertion loss can be less than 0.1 dB/facet. It can transmit almost arbitrarily many beams between two chips, without the need for single-mode waveguides or fibers—one self-aligning photonic interconnect can connect PICs with arbitrary waveguide pitches. It can be flexible and pliable and thus can conform to the surface of a PIC.
- a self-aligning photonic interconnect can be made using low-cost manufacturing techniques. And many two-dimensional (2D) self-aligning photonic interconnect can be layered or stacked on top of each other and molded together to enable 3D interconnect geometries.
- Self-aligning couplers can be realized in a system with first and second devices (e.g., PICs, microchiplets, and photonic circuit boards).
- the first device includes a first waveguide that guides light in a first direction in a first plane.
- the second device is in contact with the first device and includes a second waveguide with a coupling section that guides light in a second direction in a second plane parallel to the first plane. This second direction forms an angle of ⁇ with the first direction about an axis perpendicular to first plane and the second plane.
- the coupling section also overlaps the first waveguide (e.g., across a gap of about 0.5 ⁇ m to about 2.0 ⁇ m) such that at least a portion of the light couples evanescently between the first waveguide and the second waveguide.
- the first waveguide can be in a first waveguide array having a first pitch and the second waveguide can be in a second waveguide array having a second pitch different than the first pitch.
- the second waveguide array can comprise single-mode waveguides having different widths.
- the second pitch can be less than L sin ⁇ , where L is a length of the coupling section of the second waveguide.
- the coupling efficiency between the first and second waveguides is invariant to longitudinal displacement of the second waveguide with respect to the first waveguide, to transverse displacement of the second waveguide with respect to the first waveguide less than L sin ⁇ , and to angular misalignment about the axis perpendicular to the first plane and the second plane over a range of about 2° to about 15°.
- the first waveguide can be formed in a polymer layer that is disposed on a printed circuit board and that defines at least one hole for an electrical contact between the printed circuit board and the second device.
- the first waveguide can also be formed in or on a cantilever that is released from a substrate and pushes against a surface of the second device to align the coupling section of the second waveguide to the first waveguide along the axis perpendicular to first plane and the second plane.
- such a system may include a photonic circuit board comprising an array of linear polymer waveguides that are formed in a polymer layer on a printed circuit board and configured to guide light in a first direction. It may also include a PIC that is disposed on the photonic circuit board and includes a waveguide that crosses one of the linear polymer waveguides at an angle of about 2° and about 30° in a plane parallel to the polymer layer and at a distance of about 0.5 ⁇ m to about 2.0 ⁇ m from the linear polymer waveguide.
- the waveguide can be on a cantilever that is released from a substrate and pushes against a surface of the polymer layer to align the waveguide to the linear polymer waveguide in a direction roughly perpendicular to polymer layer.
- inventive self-aligning couplers include universal-pitch self-aligning photonic interconnects couplers (USPICs).
- a USPIC can be implemented with a first waveguide formed in a first PIC and a planar focusing element disposed on the first PIC, evanescently coupled to the first waveguide, and configured to image an output of the first waveguide to an input of a second waveguide.
- the second waveguide can be formed in the first PIC or in a second PIC.
- the planar focusing element can be thermally molded onto the first PIC. It can include a pair of cladding layers and a planar waveguide core layer disposed between the pair of planar cladding layers and having a parabolic edge and a straight edge opposite the parabolic edge.
- the planar focusing element can have an angular alignment tolerance about an axis normal to a plane of the planar focusing element of ⁇ 1/ ⁇ square root over (N) ⁇ .
- the first waveguide can be tapered down to a width at which the waveguide mode has the same group index as the planar reflector over half a coupling period between the first waveguide and the planar parabolic reflector.
- FIG. 1 A shows a self-aligning photonic circuit board (SAPCB), with insets showing (i) a self-aligning coupler with a bent or “hockey stick” waveguide that crosses over a straight waveguide in a polymer film layer of the SAPCB, (ii) a microchiplet integrated into a PIC, and (iii) electrical connections (bump bonds) through holes in the polymer film layer.
- SAPCB self-aligning photonic circuit board
- FIG. 1 B shows the integration of the microchiplet integrated into the PIC in greater detail and the translational invariance of the waveguides in the bent couplers that connect the waveguides in the microchiplet to the waveguides in the PIC.
- FIG. 2 A shows a self-aligned coupler modeled as evanescently coupled, overlapping waveguides.
- FIG. 2 B is a plot of theoretical power transfer between the bent coupler and straight waveguide in FIG. 2 A .
- FIG. 3 A is a plot of the effective mode indices as a function of the PIC (SiN) waveguide width and the polymer waveguide width for an example implementation.
- the two waveguide geometries should be engineered such that their modes have equal propagation constants.
- FIG. 3 B is a plot of power transfer efficiency as a function of the PIC waveguide width.
- FIG. 3 C is a plot of power transfer efficiency as a function of the PIC waveguide height.
- FIG. 3 D is a plot of power transfer efficiency as a function of the coupling gap between the PIC and polymer waveguides.
- FIG. 3 E is a plot of power transfer efficiency as a function of the wavelength.
- FIG. 3 F is a plot of power transfer efficiency as a function of the waveguide temperature.
- FIG. 3 G illustrates the field profile of the self-aligning coupler with parameters in Table I.
- the insets below show the cross-sectional field profile at different points along the self-aligning coupler.
- the solid lines indicate FDTD simulation results, while the dotted lines are fit to an analytical expression for the power transfer efficiency.
- FIG. 4 B is a plot of insertion loss (
- FIG. 5 A is a plot of transmission vs. ⁇ for a self-aligning coupler designed to interface a 640 ⁇ 300 nm InP gain microchiplet to a 500 nm ⁇ 220 nm silicon photonic waveguide.
- the strong mode confinement in both materials eliminates scattering loss at the intersection, permitting mode transfer with no insertion loss. As a result, the transmission characteristic reproduces nearly perfectly with the theory.
- FIG. 5 B is a plot of transmission efficiency as a function of wavelength for the self-aligning coupler of FIG. 5 A .
- This self-aligning coupler has a 1-dB bandwidth exceeding 230 nm.
- FIG. 7 A illustrates interfacing two photonic circuits with different waveguide pitches and process stacks.
- FIG. 7 B shows a waveguide in a first PIC crossing over a waveguide in a second PIC with little to no crosstalk.
- FIGS. 8 A and 8 B illustrate methods of making SAPCBs.
- FIG. 9 A shows a top view of a Universal-pitch Self-aligning Photonic Interconnect Coupler (USPIC) connecting arrays of waveguides on different PICs.
- USPIC Universal-pitch Self-aligning Photonic Interconnect Coupler
- FIG. 9 B shows a side view of a USPIC connecting arrays of waveguides on different PICs.
- FIG. 10 A is a side view of a tapered waveguide in a PIC that evanescently and adiabatically couples light into and/or out of a USPIC that overlaps an edge of the PIC.
- FIG. 10 B is a side view of a layer of high-index material between a PIC and a USPIC that evanescently and adiabatically couples light between the USPIC and a waveguide in the PIC.
- FIG. 10 C is a side view of a PIC polished to taper one end of a waveguide that is coupled evanescently and adiabatically couples light into and/or out of a USPIC that overlaps an edge of the PIC.
- FIG. 10 D is a side view of a PIC-based waveguide with one end tapered to a width at which the waveguide mode has the same group index as the USPIC over a length equal to the half the coupling period between the PIC-based waveguide and the USPIC.
- FIG. 11 shows that the angular alignment tolerance of a USPIC is much relaxed over approaches that involve aligning a single-mode waveguide in a PIC to a single-mode waveguide in a connector.
- FIG. 12 shows a USPIC (parabolic outline) patterned to image an array of modes on one PIC to an array of modes on another PIC.
- FIG. 13 A shows how the outputs of multiple PICs can be combined into one stack of 2D waveguides (USPICs), allowing for 3D arrays of beams.
- USPICs 2D waveguides
- FIG. 13 B shows a flexible USPIC connecting waveguides in a PIC in a cryostat to waveguides in a room-temperature PIC.
- FIG. 14 shows concatenated USPICs that relay of image of outputs from waveguides in one PIC to inputs of waveguides in another PIC.
- FIG. 15 A shows a splitter network terminating in a series of tapered waveguide couplers produce a low numerical aperture (NA) (i.e., large-diameter) beam in a 2D waveguide that comes to a focus and then couples into a second PIC.
- NA numerical aperture
- FIG. 15 B illustrates several splitter networks coupled by overlapping or intersecting 2D waveguides.
- FIG. 16 A shows PIC waveguides evanescently coupled to a 2D waveguide placed on top of PIC via adiabatic coupling, with no alignment needed.
- FIG. 16 B shows how unwanted mode coupling can be suppressed between adjacent overlapped waveguides by introducing an intentional mismatch in group index.
- FIG. 16 C shows closely spaced polymer waveguides with varying widths to prevent cross-coupling by lack of phase matching. These polymer waveguides couple to PIC elevator couplers because those are adiabatic.
- FIG. 16 D shows PIC waveguides coupling into arbitrary waveguides at the point where the group index in the tapering PIC waveguide matches the index of one of the overlapping USPIC waveguides.
- a group index mismatch prevents that tapering PIC waveguide from coupling to other nearby USPIC waveguides.
- FIGS. 17 A and 17 B show profile and plan views, respectively, of self-alignment in the out-of-plane (z) dimension by pre-tensioned cantilevers to ensure spring-loaded contact between the two surfaces.
- a self-aligning photonic circuit board can serve as a universal connector for optoelectronic system integration.
- An SAPCB unifies photonic integrated circuits, microchiplets, and electronics onto a single optoelectronic substrate.
- an SAPCB's waveguides can be made of polymer, making them easy and scalable to fabricate.
- the SAPCB can include only linear arrays of waveguides, making it far easier to manufacture.
- the SAPCB also includes or makes use of bent or self-aligning couplers that align the linear polymer waveguides to waveguides in the photonic components that populate the SAPCB, to other SAPCBs, and/or other waveguides.
- the self-aligning couplers provide a laterally invariant system agnostic to the exact location of waveguides on the photonics and exhibits high angular tolerance and arbitrarily high lateral tolerance.
- a self-aligning coupler can be designed to be robust to fabrication variation larger than current-day process tolerances, and its combined lateral and angular tolerance exceeds conventional optical coupling.
- the SAPCB allows for system integration with minimal design and alignment requirements, allowing for a diverse set of photonic components to interface with each other and potentially permitting standardization of photonic components.
- SAPCB Self-Aligning Photonic Circuit Board
- FIG. 1 A shows an example SAPCB 100 , which is made of a polymer-laminate film 104 bonded onto an electrical printed circuit board (PCB) 110 .
- PICs 120 are flip-chip bonded to the polymer film 104 , which includes a linear, closely-spaced array of single-mode waveguides 102 that carry signals among PICs 210 .
- the polymer waveguides 102 can be spaced with a pitch P that is smaller than the waveguide pitch p on the PIC 120 , which permits arbitrary placement of PICs 120 and PICs 120 with different (and almost arbitrary) waveguide pitches. It also enables the transmit port of a chip to be paired easily to a receiver chip on another without co-designing both PICs 120 with the SAPCB 100 .
- the inset (i) at lower left in FIG. 1 A illustrates a self-aligning coupler 124 that provides an efficient, board-level optical interconnect between the SAPCB 100 and a PIC 120 .
- a first portion 123 a of a bent or “hockey-stick” waveguide 122 in the PIC crosses over a portion of a waveguide 102 in the polymer film layer 104 of the SAPCB 100 at an angle ⁇ .
- the angle ⁇ is chosen to efficiently transfer optical power through the interaction of their evanescent fields.
- the first or bent portion of the waveguide 122 has a length L and bends at the angle ⁇ with respect to a second portion 123 b of the bent or “hockey-stick” waveguide 122 .
- the length L can range from 10-100 ⁇ m, and the bend angle can be up to 30° (e.g., 1°, 2°, 5°, 10°, 15°, 20°, 25°, or any other value up to about 30°).
- the second portion 123 b may be roughly parallel to the waveguides 102 in the SAPCB 100 , although this is not necessary.
- the overlapping portions of the waveguides 102 and 122 are separated in they direction by a gap g, which is small enough for light to couple evanescently from the PIC waveguide 122 to the SAPCB waveguide 102 or vice versa.
- the angled crossing makes the self-aligning coupler insensitive to in-plane displacements and permits coupling over a wide range of waveguide pitches, including different pitches in the PIC 120 and the SAPCB 100 . Additionally, crossing the two waveguides 102 and 122 at an angle eliminates any need to place PICs 120 onto the SAPCB 100 with sub-micron placement accuracy.
- Self-aligning couplers 124 provide highly efficient coupling with generous alignment tolerances between the SAPCB 100 and the PIC 120 .
- Approaches such as adiabatic coupling or edge coupling have demanding alignment precision (e.g., lateral alignment better than 5 ⁇ m).
- alignment precision e.g., lateral alignment better than 5 ⁇ m.
- PICs and substrates with adiabatic or edge couplers are often co-designed to ensure the placements of the polymer waveguides are matched to those of the PICs with micron-scale precision.
- PICs 120 and SAPCBs 100 with self-aligning couplers 124 do not have to be co-designed thanks to the angled design of the self-aligning couplers 124 .
- the angled geometry of the self-aligning coupler 124 is invariant to any longitudinal displacement ⁇ x (so long as the waveguides overlap) and any transverse displacement ⁇ y ⁇ L sin ⁇ . Additionally, the transverse displacement tolerance can be increased arbitrarily by increasing the length of the coupler L.
- Angled coupling introduces other benefits during assembly. Not only can the polymer waveguides 102 and PIC waveguides 122 be designed separately, but they can also have different pitches p and P, respectively, as shown in FIG. 1 A . No matter their respective pitches, as long as the two waveguides 102 and 122 are coarsely aligned within L sin ⁇ , they should cross at some point with little to no transmission penalty.
- the SAPCB 100 could therefore serve as an off-the-shelf, universal connector interfacing PICs 120 of different designs.
- the polymer waveguide pitch P is smaller than the waveguide pitch P on the PIC 120 , then two PICs 120 , potentially with different waveguide pitches and optical port locations, could be coupled to one another through a single mode polymer waveguide 102 .
- the polymer waveguide pitch P should be smaller than L sin ⁇ .
- the self-aligning coupler also enables simplified “pick-and-place” integration of microchiplets 130 into photonic circuits.
- Microchiplets 130 which are miniaturized photonic chips with isolated waveguides 132 , have recently been demonstrated for integrating gain or single-photon sources into PICs 120 .
- a PIC backbone 140 wavelength-division multiplexed (WDM) laser PIC
- WDM wavelength-division multiplexed
- the inset (ii) at lower right in FIG. 1 A shows how similar self-aligning couplers 134 simplify alignment and “pick-and-place” integration of microchiplets 130 into PICs.
- FIG. 1 B shows these self-aligning couplers 134 in greater detail.
- the microchiplet 130 is placed on the PIC 140 that includes arrays of parallel SiN waveguides 142 .
- the ends of parallel waveguides 132 in the microchiplet 130 cross over the ends of the SiN waveguides 142 in the PIC 142 at an angle ⁇ as shown in FIGS. 1 A and 1 B .
- the pitches of the waveguides 132 and 142 can be the same or different and are chosen so that an end of a given waveguide 132 in the microchiplet 130 overlaps at most the end of one waveguide 142 in the PIC 140 .
- Each pair of crossed waveguides 132 and 142 is separated by a gap in the z direction that is small enough for light to couple evanescently between them.
- FIG. 1 B illustrates a path (dashed, double-headed arrow) followed by light that couples between a pair of crossed waveguides 132 and 142 .
- Translating the waveguides 132 and 142 with respect to each other in the x-y plane does not change the amount of overlap 135 between the waveguides 132 and 142 .
- the shape and area of the crossed region 135 between the waveguides 132 and 142 remain constant, even if one waveguide is shifted laterally relative to the other waveguide (e.g., in the x or y direction).
- the coupling efficiency of the self-aligning coupler 134 scales with the length of overlap area, the coupling efficiency remains relatively constant even if the microchiplet 130 is not perfectly aligned with the PIC 140 in in the x-y plane.
- the SAPCB 100 can also accommodate electrical connections between the PCB 110 and PICs 120 or electronic components, such as dynamic random-access memory (DRAM) 112 , central processing units (CPUs) 114 , reconfigurable optical add-drop multiplexers (ROADMs) 116 , or other electronic components.
- DRAM dynamic random-access memory
- CPUs central processing units
- ROADMs reconfigurable optical add-drop multiplexers
- the inset (iii) at upper left in FIG. 1 A shows an electrical connection in the form of a bump bond between a copper trace 111 in or on the PCB 110 and a PIC 120 .
- a solder bump 113 bonds the copper trace 111 to an electrical contact on the PIC 120 through a hole punched through the polymer film 104 .
- FIG. 2 A shows two waveguides 102 and 122 that can be used to model the dynamics of a self-aligning coupler with coupled mode theory.
- ⁇ the coupling constant per unit length
- ⁇ the wavevector mismatch
- ⁇ the wavevector mismatch
- ⁇ At an arbitrary distance z along one of the waveguides, ⁇ remains unchanged, but ⁇ exponentially decays with the transverse offset
- tan ⁇ ⁇ e ⁇
- , where ⁇ ′ ⁇ tan ⁇ describes the decay of ⁇ with transverse offset per unit length.
- FIGS. 3 A- 3 F show the results of finite-difference time-domain simulations (Lumerical FDTD Solutions) of an example physical implementation of the SAPCB shown in FIG. 3 G using the parameters in TABLE 1 (below).
- These simulations assumed high-index single-mode polymer core SAPCB waveguides 102 in a low-index fluoropolymer cladding 104 separated by a gap g from silicon nitride (SiN) PIC waveguides 122 with silicon dioxide cladding.
- SiN is a high-index contrast waveguide platform transparent over visible and infrared wavelengths and is available in most silicon photonics and complementary metal-oxide-semiconductor (CMOS) foundries.
- CMOS complementary metal-oxide-semiconductor
- Polymer core n 1.575 Polymer cladding n 1.34 Polymer core dn/dT ⁇ 1.1 ⁇ 10 ⁇ 4 /° C. Polymer cladding dn/dT ⁇ 5 ⁇ 10 ⁇ 5 /° C. PIC waveguide core n 2 PIC waveguide cladding n 1.445 PIC core dn/dT 2.51 ⁇ 10 ⁇ 5 /° C. PIC cladding dn/dT 9.6 ⁇ 10 ⁇ 6 /° C.
- FIG. 3 A plots the effective mode indices as a function of the SiN and polymer waveguide widths.
- the waveguides should have matching propagation constants. This can be accomplished engineering the geometries of the waveguides. Mismatched waveguide propagation constants dominate fabrication-induced error.
- FIGS. 3 B- 3 F plot the effect on transmission caused by errors in, respectively, SiN width, SiN height, coupling gap g, wavelength, and temperature.
- the self-aligning coupler is remarkably robust to changes in all of these parameters, exhibiting less than 0.5 dB penalty for a ⁇ 20 nm variation in waveguide dimensions and lower than 0.3 dB excess loss for a ⁇ 100 nm change in the coupling gap g.
- it has a 1-dB optical bandwidth in excess of 180 nm, which is comparable to inverse tapers used in edge coupling, and less than 0.5 dB temperature sensitivity over a range of 80° C.
- the results in FIGS. 3 A- 3 F are from FDTD simulations of the full structure.
- FIG. 3 G shows these FDTD simulations for the parameters of TABLE 1.
- the upper portion of FIG. 3 G is a plan view of the simulated self-aligning coupler 124 , with shading illustrating the power transfer from the SiN waveguide 122 , through the self-aligning coupler 124 , and into the polymer waveguide 102 (or vice versa, depending on the propagation direction).
- Cross sectional field intensity plots along the structure are shown underneath in FIG. 3 G .
- the solid lines indicate FDTD simulation results, while the dotted lines are fit to equation for power transfer efficiency ⁇ given above.
- the waveguide intersection causes a scattering loss of about 0.2 dB at the optimal coupling angle ⁇ .
- ⁇ agrees well with the expression given above around this region and exhibits an angular alignment (3 dB) tolerance ⁇ >5 degrees.
- ⁇ rolls off slowly for ⁇ > ⁇ opt permitting modest coupling efficiencies at even large angular errors.
- This feature of self-aligning couplers greatly simplifies initial alignment and relaxes constraints on the alignment precision during packaging.
- FIG. 4 B is a plot of insertion loss (
- the scattering loss shown in FIG. 4 A results primarily from a faster-than-adiabatic transition at the waveguide intersection and increases with ⁇ as the transition into the hybridized modes becomes more abrupt. This loss is particularly significant at large ⁇ , shown in FIG. 4 B , which accounts for the discrepancy compared to theory at these angles.
- the scattering loss drops with increasing gap g, which makes the transition more adiabatic.
- Increasing the gap g introduces two tradeoffs, however: the angular tolerance ⁇ drops, and the transmission becomes more sensitive to errors in ⁇ . If higher insertion losses are acceptable, the waveguides can be coupled more strongly, which improves ⁇ .
- FIG. 4 A also shows such an example, where the gap g is decreased to 500 nm.
- FIG. 5 A is a plot of transmission versus coupling angle ⁇ for a self-aligning coupler designed to interface a 640 ⁇ 300 nm InP gain microchiplet to a 500 ⁇ 220 nm silicon photonic waveguide.
- the strong mode confinement in both materials eliminates scattering loss at the intersection, permitting mode transfer with no insertion loss.
- the transmission characteristic reproduces nearly perfectly the theoretical expression for transmission versus coupling angle given above.
- FIG. 5 B is a plot of transmission efficiency as a function of wavelength for the self-aligning coupler of FIG. 5 A . It shows that the coupler has a 1 dB bandwidth exceeding 230 nm.
- FIG. 6 is a plot of the lateral and angular alignment tolerances of self-aligning couplers with different gaps, tapered adiabatic couplers, and 10 ⁇ m inverse tapered edge couplers for the same materials platform used in FIGS. 3 A- 3 F .
- the traces in FIG. 6 are the 1 dB coupling efficiency contours in the ⁇ x- ⁇ plane for the different couplers.
- Non-perturbative approaches such as edge coupling (EC) have a fundamental tradeoff between the lateral and angular tolerances in coupling efficiency. Assuming the mode E wg ( ⁇ right arrow over (r) ⁇ ) produced when the waveguide couples into free space is Gaussian, one can calculate the mode overlap
- ⁇ E ⁇ C ⁇ ⁇ E w ⁇ g * ⁇ ( r ⁇ ) ⁇ E fiber ⁇ ( r ⁇ ) ⁇ d 3 ⁇ r ⁇ 2 ⁇ ⁇ E w ⁇ g ⁇ ( r ⁇ ) ⁇ 2 ⁇ ⁇ ⁇ E fiber ⁇ ( r ⁇ ) ⁇ 2 , with the input fiber mode E fiber ( ⁇ right arrow over (r) ⁇ ), also assumed to be Gaussian but misaligned by an angle ⁇ and transverse distance ⁇ x.
- Adiabatic couplers taper one or both waveguides to induce an avoided crossing between the two eigenmodes, which adiabatically transfers power from one waveguide to the other.
- This adiabatic transition makes an adiabatic coupler robust to variation in ⁇ , which has led to them being favored in many photonic platforms for their resilience to fabrication error. This robustness comes at the cost of alignment tolerance, however, as small lateral or angular errors render the interaction non-adiabatic, resulting in little or no power transfer.
- FIG. 6 shows the transmission penalty as a function of ⁇ x, ⁇ .
- the combined lateral and angular tolerance of a self-aligning coupler is therefore higher than conventional optical couplers.
- the self-aligning coupler achieves this performance by making use of perturbative coupling, which does not suffer from a fundamental limitation on ⁇ x ⁇ , and by being largely invariant to lateral displacements.
- This lateral tolerance is increased by choosing both waveguides in the self-aligning coupler to not be tapered, with the one tradeoff being that the effective indices of the waveguides should be matched.
- FIGS. 7 A and 7 B show two possible applications of the SAPCB 100 for system-level integration.
- FIG. 7 A illustrates interfacing two PICs with different waveguide pitches and process stacks, perhaps fabricated by different foundries (for clarity, FIG. 7 A shows only the waveguides 722 a and 722 b of these PICs).
- the off-axis intersection guarantees that both sets of waveguides 722 a , 722 b can couple into the same waveguides 102 on the SAPCB 100 .
- the two PICs may also not have the same process stack; for instance, one vendor may want a larger oxide passivation layer, resulting in a larger coupling gap (e.g., g 1 >g 2 as shown in FIG. 7 A ) to the polymer waveguide 102 .
- the vendor can address this by simply modifying the coupling angle ⁇ to preserve efficient power transfer.
- the SAPCB 100 could allow standardization of photonic components; for example, the geometry of the SAPCB 100 could be a published standard, while each PIC vendor designs the couplers to the SAPCB 100 on their own PIC depending on their process stack and application.
- FIG. 7 B demonstrates another advantage originating from the need for ⁇ 0 for efficient coupling.
- a polymer waveguide 712 carrying a signal from one PIC needs to travel over a waveguide 702 a in a second PIC with little to no crosstalk.
- the waveguide 702 a in the second PIC has a strong wavevector mismatch ⁇ with the polymer waveguide 712 , allowing for crosstalk-free transmission of signals over many photonic components on a board.
- the polymer waveguide 712 can also be engineered (e.g., with a different width along part of its length) to have a strong or close wavevector match with another waveguide 702 b in the second PIC.
- Bent waveguides for self-aligning couplers can be defined on a SAPCB, on a PIC, or both.
- Advanced photolithography processes can be used to define the geometry and angle precisely for a bent waveguide on PIC.
- Creating bent waveguides on PICs frees the SAPCB to include only linear arrays of polymer waveguides, with no bends or tapering required.
- the simple layout of the polymer waveguides could allow the SAPCB to be fabricated by fiber pulling approaches from a preform as shown in FIG. 8 A instead of more expensive lithography processes like those shown in FIG. 8 B .
- Polymer waveguides have a wide transparent window, making the SAPCB applicable to visible integrated photonics.
- USBs Universal-Pitch Self-Aligning Photonic Interconnect Couplers
- the upper cladding layer 904 is made of low-index material that can be as thin as a few microns or much thicker, depending on whether the USPIC 900 is rigid or flexible.
- the planar focusing element 902 is a slab waveguide with a curved edge (e.g., a parabolic edge) that borders the cladding material, air, or another material with a lower refractive index. It confines light to a plane between the cladding layers 904 and focuses light within that plane.
- the USPIC 900 in FIGS. 9 A and 9 B connects four single-mode waveguide modes between PICs 902 a and 920 b .
- the top view in FIG. 9 A illustrates the coupling from PIC 920 a of four beams into the 2D USPIC 900 , which is shown here with a parabolic reflector to image the beams at a 1:1 magnification to PIC 920 b , where the beams are again evanescently coupled into that PIC's waveguides.
- the side view in FIG. 9 B shows the adiabatic (slow compared to the period of the electromagnetic mode) coupling between waveguides 922 a in the first PIC 920 a and the USPIC 900 .
- the USPIC 900 sits on top of and partially overlaps both PICs 920 a and 920 b .
- the PICs 920 a and 920 b can also be on opposite sides of the USPIC 900 (i.e., top and bottom instead of both on the top side or the bottom).
- the USPIC 900 can also couple light between waveguides in different sections of the same PIC.
- a USPIC may include an integrated beam splitter to couple light into and/or out of the USPIC.
- a beam splitter can be made using a thin slit illuminated under total internal reflection so that a fraction of light crosses the slit to the other side.
- Other components suitable for integration into a USPIC include gratings, grating filters, and nanophotonic components, which can be made by patterning refractive index changes into the USPIC's waveguide layer with a spatial period, orientation, and location selected to diffract incident light.
- a USPIC can include a wavelength-dependent vertical grating coupler that couples light at some wavelengths into or out of the plane of the USPIC's reflector and transmits light at other wavelengths. This could be useful for routing WDM signals among different picks or for filter pump and probe beams.
- FIGS. 10 A- 10 D illustrate different elevator coupler configurations for coupling light between PICs and USPICs.
- FIG. 10 A shows a PIC 1020 a with an in-plane tapered waveguide 1022 a that is evanescently coupled to a 2D USPIC 900 .
- the PIC waveguide modes are brought near the top of the PIC 1020 a , then the dimensions of the PIC waveguide 1022 a are tapered so that the confined optical mode reaches ever further into the surrounding low-index material 1024 a .
- the PIC waveguide cores 1022 a may be made of silicon (refractive index of 3.5), whereas the surrounding cladding material 1024 a may be made of silica (refractive index of 1.5).
- the waveguide core(s) 1022 b buried in a PIC 1020 b can be evanescently coupled to a (SiN) ridge waveguide coupler 1026 on top of the PIC's surface plane.
- the USPIC 900 is positioned over this region to droop over the ridge waveguide coupler 1026 on top of the PIC 1020 b .
- the USPIC 900 can be thermally molded onto the PIC 1020 b so that the ridge waveguide coupler 1026 reaches into the bottom low-index layer 904 of the USPIC 900 .
- FIG. 10 C shows a PIC 1020 c with a vertically tapered waveguide 1022 c that is evanescently coupled to a 2D USPIC 900 .
- this way to realize an elevator coupler uses a vertical tapering of PIC waveguides 1022 c by, for example, polishing an edge facet 1028 on the PIC 1020 c.
- FIG. 10 D shows a PIC 1020 d with a waveguide 1022 d that ends in a tapered segment 1029 .
- This tapered segment 1029 is evanescently coupled to a USPIC 900 on top of the PIC 1020 d .
- the PIC waveguide 1022 d is tapered down to a width at which the waveguide mode has the same group index as the slab layer 902 in the 2D USPIC 900 .
- the 2D waveguide 902 in the USPIC 900 overlaps this tapered segment 1029 of the PIC waveguide 1022 d for a certain length Lit. Thanks to phase matching, the PIC waveguide mode can now couple to the 2D waveguide 902 in the USPIC 900 .
- L ⁇ is chosen so that it is exactly one half of a coupling period. This coupling could be considered the analog of evanescent directional coupling commonly used in PIC beam splitters.
- FIG. 11 illustrates the longitudinal displacement tolerance ⁇ z and angular alignment tolerance ⁇ for coupling light between a USPIC 900 and one or more waveguides 1022 in a PIC 1020 .
- a USPIC has a much larger tolerance to misalignment than other types of couplers.
- NA numerical aperture
- FIG. 12 shows a USPIC 1200 that couples light between a pair of PICs 1220 a and 1220 b .
- This USPCI 1200 has a straight edge 1201 that overlaps the top of the PICs 1220 a and 1220 b ; on the opposing side of the USPIC 1200 is a parabolic edge 1203 that serves as a parabolic reflector.
- the PICs 1220 a and 1220 b are positioned at a distance from the USPIC's center axis selected to ensure that the beams from the waveguides undergo total internal reflection at the boundary defined by the parabolic edge 1203 of the USPIC 1200 .
- the rays in FIG. 12 trace the paths taken by the beams.
- the beams diverge in the plane of the 2D waveguide layer of the USPIC 1200 .
- the parabolic reflector refocuses them in its image plane, e.g., with 1:1 imaging as in FIG. 12 .
- Magnification and demagnification between PICs are also possible by choosing the appropriate distances along the imaging axis of the USPIC 1200 .
- the beams propagating in the USPIC 1200 are not transversely confined by patterned ridges (as in waveguides typical in PICs), there are no edge roughness losses except perhaps at the point of reflection from the parabolic edge 1203 (which can be made very smooth as mentioned below).
- the top and bottom of the waveguide layer (e.g., layer 902 in FIG. 9 B ) in the USPIC 1200 can also be made very smooth (see below), so the primary loss may be due to absorption in the waveguiding material itself. In the case of SU-8 or PMMA, for example, these losses can be below 0.1 dB/cm in a wide wavelength range from blue to near infrared.
- a low-index polymer e.g., PMMA
- a rigid carrier e.g., PMMA
- a high-index polymer such as SU-8
- optical lithography may be used to pattern the high-index layer into the desired shape(s), e.g., with a parabolic curved edge; alternatively, the entire stack can be patterned later.
- the high-index layer is dried and optionally annealed, then capped with another low-index polymer. These fabrication steps can be repeated to create multiple 2D waveguide layers for edge-coupling multiple chips, where the evanescent couplers are tiered (for example, using the edge facet tapered waveguide coupling of FIG. 13 A , described below).
- a USPIC can also be made from a pre-form with a high-index layer sandwiched between a pair of low-index layers.
- the pre-form can have many alternating high- and low-index layers.
- These layers are several times thicker than ultimately desired.
- the pre-form is patterned into the desired shape (e.g., with a parabolic or curved edge opposite a straight edge), then stretched to the desired film thickness under heating to produce the USPIC.
- the pre-form can be patterned using optical lithography, cutting, or molding. Alternatively, the pre-form can be stretched, then cut and/or patterned into the desired shape.
- additional thermal reflow steps can be used during fabrication.
- the sample can be heated just above the melting point of the waveguiding material so that surface tension reduces out surface roughness.
- FIG. 13 A shows four separate PICs 1320 a - 1320 d arrayed vertically, with parallel N-beam waveguide arrays arranged into and/or out of the page.
- the PICs 1320 a - 1320 d are coupled to a USPIC stack 1300 via separate bent, flexible USPIC waveguide layers 1302 a - 1302 d .
- This forms a pitch-reducing optical beam array that can be coupled to or formed into a 2D array of waveguides, where each waveguide can be connected to a PIC component, such as a modulator.
- 2D beam arrays at high speed or high power are not currently available and are useful for applications such as holography, beam steering/lidar, optical control of cold atoms or color centers, endoscopy, optical communications, etc.
- FIG. 13 B shows a flexible USPIC 1350 connecting a cryogenically cooled PIC 1370 a (e.g., at temperatures of 77 K or lower) on a stage 1392 in a cryostat 1390 to a PIC 1370 b at room temperature (e.g., at a temperature of 285-300 K).
- the USPIC 1350 can be used as flexible photonic ribbon cables interfacing the PICs 1370 a and 1370 b .
- This flexible photonic ribbon cable 1350 enables flexible interconnects between photonic subsystems, much like electrical ribbon cables are used for connecting electrical subsystems. Other photonic ribbon cables are not self-aligning and are thus more difficult to align and maintain.
- the self-aligning USPIC 1350 solves that problem and therefore provides easy to align, reliable photonic interconnects between components inside and outside the cryostat 1390 .
- FIG. 14 shows a relay 1400 of 2D imaging USPICs that bridges a longer distance between a pair of waveguide arrays 1420 a and 1420 b .
- This relay 1400 includes multiple USPICs, merged together in the same plane, with curved (e.g., parabolic or circular) edges that face in opposite directions to image the output facets of waveguides in PIC 1420 a to the input facets of waveguides in PIC 1420 b via intermediate image planes 1401 .
- curved e.g., parabolic or circular
- the imaging is 1:1 (i.e., unity magnification), so the curved edges have identical shapes.
- the curved edges can also have different shapes or radii of curvature to magnify or demagnify the images of the waveguide facets, e.g., to accommodate waveguide arrays with different pitches.
- the relay 1400 can have more or fewer curved edges, depending on the distance separating the two PICs 1420 a and 1420 b and the orientation of the waveguides in the PICs 1420 a and 1420 b with respect to each other.
- FIGS. 15 A and 15 B show USPICs without 2D Focusing elements.
- a waveguide in a PIC 1520 a can be fanned out to an array of tapered waveguides 1522 a arranged to emit a focusing wave as in a phased array (considered differently, diverging components of beams are coherently cancelled).
- These tapered waveguides 1522 a are evanescently coupled to a USPIC 1500 with a slab waveguide layer that bridges a connection to tapered waveguides 1522 b in another PIC 1520 b .
- the USPIC 1500 does not need a focusing element (such as parabolic reflector).
- the same technique can be applied to bridge many optical channels, as illustrated in FIG. 15 B .
- FIGS. 16 A- 16 D illustrate how to interconnect PICs 1620 a and 6120 b using a USPIC 1600 with a closely (e.g., sub-wavelength) spaced array of waveguides 1602 .
- the USPIC acts as a “ribbon cable” that connects the PICs 1620 a and 1620 b at an arbitrary distance, since beams 1601 are prevented from spreading between waveguides by group velocity mismatch (group index mismatch/lack of phase-matching) among the waveguides 1602 .
- FIG. 16 B is a plot of the group index versus transverse position for the waveguides 1602 , with each horizontal line segment representing a different waveguide.
- the vertical offsets indicate the group index (and hence group velocity) mismatches.
- the beams 1601 in the USPIC waveguides 1602 couple adiabatically to and/or form waveguides 1622 in each PIC 1620 a , 1620 b via elevator coupling (i.e., the waveguides 1602 and 1622 have tapered ends that are overlapped vertically).
- This USPIC 1620 has patterned waveguides instead of the 2D waveguide described above.
- a limitation of this approach is that it does not allow for crossing beams as in the relay 1400 shown in FIG. 14 or slab waveguide USPIC 1500 in FIG. 15 .
- a benefit of this approach is that arbitrary lengths are possible without intermediate refocusing of the beams as above (e.g., as in FIG. 14 ).
- Scalable and low-cost manufacture is still possible by, for example, pulling pre-forms from a polymer ingot in a draw-tower for increased uniformity. In this approach, angular misalignment is still acceptable thanks to the evanescent adiabatic coupling described above.
- FIG. 16 D shows coupling between waveguides 1602 in the USPIC 1600 and the waveguides 1622 in the first PIC 1622 a in greater detail.
- the waveguides 1602 , 1622 have different pitches and can tolerate a wide range of angular misalignment, just like the waveguide coupling between the PIC and microchiplet in FIG. 1 B . Indeed, an intentional angular offset can actually be beneficial in this approach.
- Off-axis alignment has at least two benefits. First, the system has improved lateral tolerance when the waveguide intersects the coupler at an angle. Second, off-axis coupling tends to improve angular tolerance. Both of these benefits are described above in greater detail.
- FIGS. 17 A and 17 B show profile and plan views, respectively, or self-aligning couplers 1724 that provide automatic alignment in three dimensions (3D), i.e., in the transverse plane and in the vertical dimension, between waveguides 1722 a and 1722 b in PICs 1720 a and 1720 b , respectively.
- 3D alignment is useful because whereas the in-plane self-alignment works well when the surfaces of the two waveguide-containing materials are flat so that they can be positioned well on top of one another, flatness is not always guaranteed.
- the upper PIC 1720 b has an array of parallel SiN waveguides 1722 b formed on an SiO 2 cladding layer 1726 b , which is in turn is on a suitable substrate 1728 b .
- Its surface is nominally flat, but could be rippled, curved, or bumpy, e.g., due to manufacturing imperfections or to meet specified design criteria.
- the lower PIC 1720 a also includes an array of SiN waveguides 1722 a formed on an SiO 2 cladding layer 1726 a , which is in turn is on a suitable substrate 1728 a .
- the SiN waveguides 1722 a are parallel with each other but are formed in a bent or hockey-stick shape when viewed from above ( FIG. 17 B ) like the couplers shown in inset (i) of FIG. 1 A . (They can also be straight but angled with respect to the waveguides 1722 b in the upper PIC 1720 b like the microchiplet and PIC waveguides in inset (ii) of FIGS. 1 A and 1 n FIG. 1 B .)
- the bent portions of the waveguides 1722 a and a portion of the underlying cladding 1726 a layer form a cantilever 1725 that is released from the substrate 1728 a .
- a trench or undercut region 1727 is etched in the substrate 1728 a underneath the cantilever using an isotropic chemical etch for a selective release like those used in microelectromechanical systems (MEMS) manufacturing. Because the SiN waveguide layer tends to be strained in tension, releasing it curls the oxide membrane up and toward the upper PIC 1720 b (out of the page in FIG. 17 B ).
- a cantilever 1725 that is about 50 ⁇ m long and includes 300 nm thick SiN waveguide layer on a 1 ⁇ m thick SiO 2 layer should reach out of the plane of the lower PIC 1720 a by about 10-20 ⁇ m. As expected for a singly clamped bimorph layer, the out-of-plane displacement grows quadratically with cantilever length, provided the displacement is small enough to keep it from curling up.
- FIG. 17 A shows the benefit of this cantilever 1725 a clearly: when the PICs 1720 a and 1720 b are stacked on each other, the cantilever 1725 comes to rest spring-loaded against the upper PIC 1720 b .
- the alignment between the waveguides 1722 a and 1722 b is resilient to modest translations in all three dimensions (x, y, and z).
- the same spring-loaded vertical alignment can be used for electrical contacts and for contacts against a self-aligning photonic circuit board (e.g., as in FIG. 1 A ) or self-aligning ribbon cable (e.g., as in FIGS. 16 A- 16 D ).
- the cantilever 1725 can be singly clamped, as illustrated in FIGS. 17 A and 17 B , or doubly clamped for stiffer, but smaller, out of plane displacement.
- the cantilever 1725 and waveguides 1722 a and 1722 b are designed so that the waveguides 1722 a and 1722 b overlap, intersect, and/or cross over each other at a single, uniquely shaped region.
- This intersection is designed (by waveguide width, height, and intersection angle) such that light 1721 propagating in one set of waveguides 1722 a ( b ) evanescently couples into the other set of waveguides 1722 b ( a ).
- the amount of coupling depends on the size and shape of the intersection and can be targeted at 100% transmission between the layers, or some fraction. This coupling technique can be used between PICs, as shown in FIGS.
- the waveguides on the cantilever can be bent or straight; if they are straight, the waveguides on the other device (e.g., the second PIC 1720 b in FIG. 17 A ) can be bent or angled to provide tolerance to transverse misalignment.
- any combination of two or more such features, systems, articles, materials, kits, and/or methods, if such features, systems, articles, materials, kits, and/or methods are not mutually inconsistent, is included within the inventive scope of the present disclosure.
- Other substitutions, modifications, changes, and omissions may be made in the design, operating conditions and arrangement of respective elements of the implementations without departing from the scope of the present disclosure.
- the use of a numerical range does not preclude equivalents that fall outside the range that fulfill the same function, in the same way, to produce the same result.
- a reference to “A and/or B”, when used in conjunction with open-ended language such as “comprising” can refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.
- the phrase “at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements.
- This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified.
- “at least one of A and B” can refer, in one embodiment, to at least one, optionally including more than one, A, with no B present (and optionally including elements other than B); in another embodiment, to at least one, optionally including more than one, B, with no A present (and optionally including elements other than A); in yet another embodiment, to at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc.
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Microelectronics & Electronic Packaging (AREA)
- General Physics & Mathematics (AREA)
- Optics & Photonics (AREA)
- Optical Integrated Circuits (AREA)
Abstract
Description
The dimensions of the two waveguides are chosen to ensure that their effective mode indices nPIC, nSAPCB are equal, i.e., Δ=2π (nSAPCB−nPIC)/λ=0. Analytically solving these equations provides the power transfer efficiency η at an angle θ when the effective indices are matched:
The power transfer efficiency η reaches unity when the argument of the sine function is π/2, i.e., θopt=arctan[(4/π)κ/γ]. The 3-dB angular tolerance Δθ is therefore:
where tan−1 θ≈θ for small coupling angles θ.
-
- High angular tolerance: The 1/tan θ dependence of η produces a large angular tolerance Δθ=(4/3)θopt. Moreover, η has a long tail that ensures modest coupling even at very large angular errors, greatly simplifying initial alignment. Coupling the waveguides more strongly (increasing κ/γ) further increases Δθ.
- Robust design: no matter then values of κ, γ, the coupling efficiency reaches unity at some angle. This suggests that fabrication-induced variation in κ can therefore always be corrected during alignment. No matter the design, the angled coupler allows efficient power transfer by rotating one waveguide relative to the other. By contrast, errors in κ from the designed value reduce the efficiency of conventional adiabatic and directional couplers. It may not be possible to correct these errors in conventional adiabatic and directional couplers after fabrication.
Simulated Performance of Self-Aligning Couplers
TABLE 1 |
Simulation Parameters |
Polymer core n | 1.575 | |
Polymer cladding n | 1.34 | |
Polymer core dn/dT | −1.1 × 10−4/° C. | |
Polymer cladding dn/dT | −5 × 10−5/° C. | |
PIC |
2 | |
PIC waveguide cladding n | 1.445 | |
PIC core dn/dT | 2.51 × 10−5/° C. | |
PIC cladding dn/dT | 9.6 × 10−6/° C. |
SiN width | 462.5 | nm | |
Polymer width | 1.6 | μm | |
SiN height | 300 | | |
Polymer height | |||
1 | μm | ||
Gap (g) | 1 | μm | |
Length (L) | 100 | μm |
θopt | 4.4° |
Wavelength (λ) | 1550 | nm | ||
with the input fiber mode Efiber({right arrow over (r)}), also assumed to be Gaussian but misaligned by an angle δθ and transverse distance δx. Assume that Efiber({right arrow over (r)}) and Ewg({right arrow over (r)}) have identical beam waist radius w0; therefore, ηEC is unity when there is no misalignment. For an angular error δθ in the paraxial limit, the absolute value of the mode overlap is:
This translates to a fundamental tradeoff between the 3 dB lateral (Δx) and angular (Δθ) tolerances:
Claims (20)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US17/470,803 US12038608B2 (en) | 2020-11-09 | 2021-09-09 | Self-aligning photonic interconnections for photonic integrated circuits |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US202063111232P | 2020-11-09 | 2020-11-09 | |
US17/470,803 US12038608B2 (en) | 2020-11-09 | 2021-09-09 | Self-aligning photonic interconnections for photonic integrated circuits |
Publications (2)
Publication Number | Publication Date |
---|---|
US20220146749A1 US20220146749A1 (en) | 2022-05-12 |
US12038608B2 true US12038608B2 (en) | 2024-07-16 |
Family
ID=81453406
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US17/470,803 Active 2042-08-10 US12038608B2 (en) | 2020-11-09 | 2021-09-09 | Self-aligning photonic interconnections for photonic integrated circuits |
Country Status (1)
Country | Link |
---|---|
US (1) | US12038608B2 (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR3116913B1 (en) * | 2020-11-30 | 2022-12-16 | Commissariat Energie Atomique | Assembly comprising a first and a second photonic chips placed on top of each other |
US12066671B2 (en) * | 2022-01-12 | 2024-08-20 | Taiwan Semiconductor Manufacturing Company, Ltd. | Semiconductor devices with vertically stacked and laterally offset intermediate waveguides |
US11892680B2 (en) * | 2022-06-29 | 2024-02-06 | Globalfoundries U.S. Inc. | Edge couplers with a high-elevation assistance feature |
DE102023202509A1 (en) | 2023-03-21 | 2024-09-26 | Robert Bosch Gesellschaft mit beschränkter Haftung | Photonic system and method for producing a photonic system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8948555B1 (en) * | 2013-05-21 | 2015-02-03 | Aurrion, Inc. | Skew waveguide directional coupler |
US20170139142A1 (en) * | 2015-11-13 | 2017-05-18 | Cisco Technology, Inc. | Photonic chip with an evanescent coupling interface |
US20170343734A1 (en) * | 2015-05-05 | 2017-11-30 | Huawei Technologies Co., Ltd. | Optical coupling arrangement |
US20180172905A1 (en) * | 2016-12-21 | 2018-06-21 | Corning Optical Communications LLC | Flexible glass optical-electrical interconnection device and photonic assemblies using same |
US20200225401A1 (en) | 2019-01-15 | 2020-07-16 | Shaoliang YU | Integrated Freeform Optical Couplers |
-
2021
- 2021-09-09 US US17/470,803 patent/US12038608B2/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8948555B1 (en) * | 2013-05-21 | 2015-02-03 | Aurrion, Inc. | Skew waveguide directional coupler |
US20170343734A1 (en) * | 2015-05-05 | 2017-11-30 | Huawei Technologies Co., Ltd. | Optical coupling arrangement |
US20170139142A1 (en) * | 2015-11-13 | 2017-05-18 | Cisco Technology, Inc. | Photonic chip with an evanescent coupling interface |
US20180172905A1 (en) * | 2016-12-21 | 2018-06-21 | Corning Optical Communications LLC | Flexible glass optical-electrical interconnection device and photonic assemblies using same |
US20200225401A1 (en) | 2019-01-15 | 2020-07-16 | Shaoliang YU | Integrated Freeform Optical Couplers |
Non-Patent Citations (35)
Also Published As
Publication number | Publication date |
---|---|
US20220146749A1 (en) | 2022-05-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US12038608B2 (en) | Self-aligning photonic interconnections for photonic integrated circuits | |
US9817193B2 (en) | Through-substrate optical coupling to photonics chips | |
US10698164B2 (en) | Optical apparatus and methods of manufacture thereof | |
Kopp et al. | Silicon photonic circuits: on-CMOS integration, fiber optical coupling, and packaging | |
US9405066B2 (en) | Two-stage adiabatically coupled photonic systems | |
US6987913B2 (en) | Optical junction apparatus and methods employing optical power transverse-transfer | |
KR101333418B1 (en) | A polarization diversity grating coupler comprising a two dimensional grating | |
US8787712B2 (en) | Low loss directional coupling between highly dissimilar optical waveguides for high refractive index integrated photonic circuits | |
Barwicz et al. | Automated, high-throughput photonic packaging | |
WO2016131355A1 (en) | Photonic chip surface grating coupler (sgc) -based optical splitter and optical combiner | |
US20150331197A1 (en) | Method And System For Grating Couplers Incorporating Perturbed Waveguides | |
US10802215B2 (en) | Optical waveguide, corresponding coupling arrangement, apparatus and method | |
KR20150024431A (en) | Optical assembly with diffractive optical element | |
US10809456B2 (en) | Adiabatically coupled photonic systems with fan-out interposer | |
US9791640B2 (en) | Interposer with separable interface | |
US11067754B2 (en) | Simultaneous electrical and optical connections for flip chip assembly | |
US20200132931A1 (en) | Relaxed tolerance adiabatic coupler for optical interconnects | |
Bandyopadhyay et al. | Alignment-free photonic interconnects | |
Mangal et al. | Expanded-beam backside coupling interface for alignment-tolerant packaging of silicon photonics | |
Barwicz et al. | Breaking the mold of photonic packaging | |
US11886014B2 (en) | Two-stage expanded beam optical coupling | |
US12140795B2 (en) | Optical 2D spot-size conversion | |
Palen | Optical coupling to monolithic integrated photonic circuits | |
US20230030105A1 (en) | Optical 2D spot-size conversion | |
CN117270101A (en) | Coarse wavelength division multiplexer/demultiplexer and photon integrated chip |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
FEPP | Fee payment procedure |
Free format text: ENTITY STATUS SET TO UNDISCOUNTED (ORIGINAL EVENT CODE: BIG.); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY |
|
FEPP | Fee payment procedure |
Free format text: ENTITY STATUS SET TO SMALL (ORIGINAL EVENT CODE: SMAL); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: DOCKETED NEW CASE - READY FOR EXAMINATION |
|
AS | Assignment |
Owner name: MASSACHUSETTS INSTITUTE OF TECHNOLOGY, MASSACHUSETTS Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:BANDYOPADHYAY, SAUMIL;ENGLUND, DIRK ROBERT;SIGNING DATES FROM 20211210 TO 20220102;REEL/FRAME:059016/0023 |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: NON FINAL ACTION MAILED |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: RESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINER |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: NON FINAL ACTION MAILED |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: RESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINER |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: NOTICE OF ALLOWANCE MAILED -- APPLICATION RECEIVED IN OFFICE OF PUBLICATIONS |
|
ZAAB | Notice of allowance mailed |
Free format text: ORIGINAL CODE: MN/=. |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: AWAITING TC RESP., ISSUE FEE NOT PAID |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: NOTICE OF ALLOWANCE MAILED -- APPLICATION RECEIVED IN OFFICE OF PUBLICATIONS |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: PUBLICATIONS -- ISSUE FEE PAYMENT RECEIVED |
|
STPP | Information on status: patent application and granting procedure in general |
Free format text: PUBLICATIONS -- ISSUE FEE PAYMENT VERIFIED |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |